Add initial support for monitoring GPUs on Linux#1998
Merged
SuperQ merged 1 commit intoprometheus:masterfrom Oct 11, 2021
siavashs:drm
Merged
Add initial support for monitoring GPUs on Linux#1998SuperQ merged 1 commit intoprometheus:masterfrom siavashs:drm
SuperQ merged 1 commit intoprometheus:masterfrom
siavashs:drm
Conversation
Contributor
Author
|
Sample metrics from a Vega64: # HELP node_drm_card_info Card information
# TYPE node_drm_card_info gauge
node_drm_card_info{card="card0",memory_vendor="samsung",power_performance_level="manual",unique_id="1234567890",vendor="amd"} 1
# HELP node_drm_gpu_busy_percent How busy the GPU is as a percentage.
# TYPE node_drm_gpu_busy_percent gauge
node_drm_gpu_busy_percent{card="card0"} 10
# HELP node_drm_memory_gtt_size_bytes The size of the graphics translation table (GTT) block in bytes.
# TYPE node_drm_memory_gtt_size_bytes gauge
node_drm_memory_gtt_size_bytes{card="card0"} 8.573157376e+09
# HELP node_drm_memory_gtt_used_bytes The used amount of the graphics translation table (GTT) block in bytes.
# TYPE node_drm_memory_gtt_used_bytes gauge
node_drm_memory_gtt_used_bytes{card="card0"} 1.48447232e+08
# HELP node_drm_memory_vis_vram_size_bytes The size of visible VRAM in bytes.
# TYPE node_drm_memory_vis_vram_size_bytes gauge
node_drm_memory_vis_vram_size_bytes{card="card0"} 2.68435456e+08
# HELP node_drm_memory_vis_vram_used_bytes The used amount of visible VRAM in bytes.
# TYPE node_drm_memory_vis_vram_used_bytes gauge
node_drm_memory_vis_vram_used_bytes{card="card0"} 1.13287168e+08
# HELP node_drm_memory_vram_size_bytes The size of VRAM in bytes.
# TYPE node_drm_memory_vram_size_bytes gauge
node_drm_memory_vram_size_bytes{card="card0"} 8.573157376e+09
# HELP node_drm_memory_vram_used_bytes The used amount of VRAM in bytes.
# TYPE node_drm_memory_vram_used_bytes gauge
node_drm_memory_vram_used_bytes{card="card0"} 1.773531136e+09 |
Contributor
Author
|
FreeBSD supports the same Linux driver but I'm not sure if it exposes the DRM information through sysfs. |
Member
|
Nice! But we should move the parsing to https://github.com/prometheus/procfs - Can you submit a PR there? That'd be great! |
Contributor
Author
|
@discordianfish prometheus/procfs#370 |
Contributor
Author
|
@SuperQ when will the next |
Contributor
Author
|
@discordianfish this is refactored and ready for review 😄 |
Contributor
Author
|
I think the only required change would be to set the |
Expose GPU metrics using `sysfs/drm`. `amdgpu` is the only driver which exposes this information through DRM. Signed-off-by: Siavash Safi <siavash.safi@gmail.com>
|
Any update on this? |
Member
|
@SuperQ Can you take a look? |
SuperQ
added a commit
that referenced
this pull request
Oct 18, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
SuperQ
added a commit
that referenced
this pull request
Oct 19, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
Merged
SuperQ
added a commit
that referenced
this pull request
Oct 27, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [CHANGE] Exclude filesystems under /run/credentials #2157
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add Darwin thermal collector #2032
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add DMI collector #2131
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector #2169
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
SuperQ
added a commit
that referenced
this pull request
Oct 28, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [CHANGE] Exclude filesystems under /run/credentials #2157
* [FEATURE] Add darwin powersupply collector #1777
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add Darwin thermal collector #2032
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add DMI collector #2131
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector #2169
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
SuperQ
added a commit
that referenced
this pull request
Oct 29, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [CHANGE] Exclude filesystems under /run/credentials #2157
* [FEATURE] Add lnstat collector for metrics from /proc/net/stat/ #1771
* [FEATURE] Add darwin powersupply collector #1777
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add Darwin thermal collector #2032
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add DMI collector #2131
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector #2169
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
SuperQ
added a commit
that referenced
this pull request
Nov 18, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [CHANGE] Exclude filesystems under /run/credentials #2157
* [FEATURE] Add lnstat collector for metrics from /proc/net/stat/ #1771
* [FEATURE] Add darwin powersupply collector #1777
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add Darwin thermal collector #2032
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add DMI collector #2131
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector #2169
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
SuperQ
added a commit
that referenced
this pull request
Nov 18, 2021
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector #2146
* [CHANGE] Exclude filesystems under /run/credentials #2157
* [FEATURE] Add lnstat collector for metrics from /proc/net/stat/ #1771
* [FEATURE] Add darwin powersupply collector #1777
* [FEATURE] Add support for monitoring GPUs on Linux #1998
* [FEATURE] Add Darwin thermal collector #2032
* [FEATURE] Add os release collector #2094
* [FEATURE] Add netdev.address-info collector #2105
* [ENHANCEMENT] Support glob textfile collector directories #1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric #2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering #2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics #2123
* [ENHANCEMENT] Add DMI collector #2131
* [ENHANCEMENT] Add threads metrics to processes collector #2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector #2169
* [BUGFIX] ethtool: Sanitize metric names #2093
* [BUGFIX] Fix ethtool collector for multiple interfaces #2126
* [BUGFIX] Fix possible panic on macOS #2133
* [BUGFIX] Collect flag_info and bug_info only for one core #2156
Signed-off-by: Ben Kochie <superq@gmail.com>
oblitorum
pushed a commit
to shatteredsilicon/node_exporter
that referenced
this pull request
Apr 9, 2024
Expose GPU metrics using `sysfs/drm`. `amdgpu` is the only driver which exposes this information through DRM. Signed-off-by: Siavash Safi <siavash.safi@gmail.com>
oblitorum
pushed a commit
to shatteredsilicon/node_exporter
that referenced
this pull request
Apr 9, 2024
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector prometheus#2146
* [CHANGE] Exclude filesystems under /run/credentials prometheus#2157
* [FEATURE] Add lnstat collector for metrics from /proc/net/stat/ prometheus#1771
* [FEATURE] Add darwin powersupply collector prometheus#1777
* [FEATURE] Add support for monitoring GPUs on Linux prometheus#1998
* [FEATURE] Add Darwin thermal collector prometheus#2032
* [FEATURE] Add os release collector prometheus#2094
* [FEATURE] Add netdev.address-info collector prometheus#2105
* [ENHANCEMENT] Support glob textfile collector directories prometheus#1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric prometheus#2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering prometheus#2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics prometheus#2123
* [ENHANCEMENT] Add DMI collector prometheus#2131
* [ENHANCEMENT] Add threads metrics to processes collector prometheus#2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector prometheus#2169
* [BUGFIX] ethtool: Sanitize metric names prometheus#2093
* [BUGFIX] Fix ethtool collector for multiple interfaces prometheus#2126
* [BUGFIX] Fix possible panic on macOS prometheus#2133
* [BUGFIX] Collect flag_info and bug_info only for one core prometheus#2156
Signed-off-by: Ben Kochie <superq@gmail.com>
oblitorum
pushed a commit
to shatteredsilicon/node_exporter
that referenced
this pull request
Apr 9, 2024
Expose GPU metrics using `sysfs/drm`. `amdgpu` is the only driver which exposes this information through DRM. Signed-off-by: Siavash Safi <siavash.safi@gmail.com>
oblitorum
pushed a commit
to shatteredsilicon/node_exporter
that referenced
this pull request
Apr 9, 2024
NOTE: In order to support globs in the textfile collector path, filenames exposed by
`node_textfile_mtime_seconds` now contain the full path name.
* [CHANGE] Add path label to rapl collector prometheus#2146
* [CHANGE] Exclude filesystems under /run/credentials prometheus#2157
* [FEATURE] Add lnstat collector for metrics from /proc/net/stat/ prometheus#1771
* [FEATURE] Add darwin powersupply collector prometheus#1777
* [FEATURE] Add support for monitoring GPUs on Linux prometheus#1998
* [FEATURE] Add Darwin thermal collector prometheus#2032
* [FEATURE] Add os release collector prometheus#2094
* [FEATURE] Add netdev.address-info collector prometheus#2105
* [ENHANCEMENT] Support glob textfile collector directories prometheus#1985
* [ENHANCEMENT] ethtool: Expose node_ethtool_info metric prometheus#2080
* [ENHANCEMENT] Use include/exclude flags for ethtool filtering prometheus#2165
* [ENHANCEMENT] Add flag to disable guest CPU metrics prometheus#2123
* [ENHANCEMENT] Add DMI collector prometheus#2131
* [ENHANCEMENT] Add threads metrics to processes collector prometheus#2164
* [ENHANCMMENT] Reduce timer GC delays in the Linux filesystem collector prometheus#2169
* [BUGFIX] ethtool: Sanitize metric names prometheus#2093
* [BUGFIX] Fix ethtool collector for multiple interfaces prometheus#2126
* [BUGFIX] Fix possible panic on macOS prometheus#2133
* [BUGFIX] Collect flag_info and bug_info only for one core prometheus#2156
Signed-off-by: Ben Kochie <superq@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Expose GPU metrics using
sysfs/drm.amdgpuis the only driver which exposes this information through DRM.